PFM-1: Visible and Near-Infrared Hyperspectral Imagery for Landmine Detection

Visible and Near-Infrared Hyperspectral Imaging data prepared for the IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2026. This specific version is a refined subset of an original benchmark dataset. The dataset was authored by SagarLekhak and last updated on March 22, 2026.

Use Cases

• Train image classification models for landmine detection based on hyperspectral signatures.
• Develop image segmentation algorithms to identify landmine locations from UAV-based imagery.
• Benchmark hyperspectral data processing pipelines for geospatial analysis tasks.
• Refine or validate models using a curated subset of a larger benchmark dataset.

Strengths

• Dataset is prepared for a major academic conference (IEEE IGARSS 2026).
• Data is a refined subset of an original benchmark dataset, suggesting some curation.
• Last update timestamp is explicitly provided (2026-03-22 18:37:05).

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license are unknown, which may limit suitability assessment.

Provenance

Source

huggingface

Collection Method

Likely collected via UAV (Unmanned Aerial Vehicle) with a VNIR hyperspectral sensor.

Time Range

null

Freshness

Last updated 2026-03-22 18:37:05; freshness should be verified.

Geography

Region is tagged as 'us', but specific spatial coverage is unknown.